Mycobacterium tuberculosis Genes Proof of Concept Experimentation

In the late 1980s, health officials first detected drug-resistant tuberculosis (TB) strains, prompting the World Health Organization (WHO) to classify TB as a global health emergency (WHO, 2023). Since then, the situation has worsened, marked by the emergence of multidrug-resistant tuberculosis (MDR-TB) strains. Antimicrobial resistance (AMR) is particularly prevalent in TB infections, with 78% of the 500,000 rifampicin-resistant (RR) cases in 2020 classified as multidrug-resistant. TB remains the deadliest single pathogen globally, causing significant mortality with nearly 10 million cases and 1.5 million deaths annually (Ichsan et al., 2023). Consequently, Lambert iGEM chose to demonstrate the proposed approach of using CRISPRi to downregulate the critical inhA gene in Mycobacterium tuberculosis (M. tb) as a proof of concept for the SHIELD toolbox.

inhA

Among M. tuberculosis genes, we specifically chose to target the inhA gene due to its low mutation rate and its critical role in M. tuberculosis’s pathogenicity and survival, making it an ideal proof of concept for testing our constructs. inhA encodes the NADH-dependent enoyl-acyl carrier protein reductase that synthesizes type II fatty acids, essential components of mycolic acids in the mycobacterial cell envelope linked to M. tuberculosis’s virulence (Marrakchi et al., 2014). By inhibiting mycolic acid synthesis with our CRISPRi system, we aim to disrupt M. tuberculosis’s pathogenicity and eliminate the bacteria. This approach, targeting an essential gene with low mutation rates using the dCas9-sgRNA complex, enhances gene downregulation and disrupts cell wall stability, reducing the bacteria’s virulence.

Design

inhA sgRNAs

We retrieved the full inhA gene sequence from the National Library of Medicine and used Benchling’s software to identify sgRNA binding sites that maximized on-target effects (Benchling, 2023) (see Fig. 1). This process allowed us to identify four optimal sgRNA sequences predicted to achieve the highest on-target effects on the non-coding strand of the inhA gene sequence for the CRISPRi system (see Fig. 2).

Figure 1. Diagram of the target locations of sgRNAs on the non-coding strand of the inhA gene sequence.

Figure 2. Four optimal inhA sgRNA target sequence generated from Benchling.

After selecting our sgRNA target sequence, we substituted these sequences in place of the GFP targeting sgRNAs, sgRNA9 (BBa_K5096074) and sgRNA6 (BBa_K5096620) (see CRISPRi GFP), within the template sgRNA constructs as detailed in the supplementary materials of Marshall et al. (2020) (see Fig. 3). We then ordered our four inhA sgRNAs as linear gBlocks from Integrated DNA Technologies (IDT).

Figure 3. Linear DNA construct of inhA sgRNA, modified from Marshall et al. (2020) by substituting the original sequence with the target inhA sequence. Forward and reverse primers (amplify_gBlock) added to the front and back of the sgRNA sequence.

inhA Target Construct

We designed our inhA target (BBa_K5096046) construct by assembling a linear DNA sequence that incorporated primers (pBEST-seq6-s and pBEST-seq6-as), T7 promoter, or2-or1 promoter, inhA sequences complementary to the target sgRNAs, UTR1 (RBS), deGFP, T7 terminator, and or2-or1 terminator (see Fig. 4). We placed the deGFP gene after the inhA sequence, as successful binding of sgRNAs to the inhA gene sequence would inhibit downstream GFP expression, resulting in a quantifiable decrease in fluorescence from the inhA CRISPRi reaction.

Figure 4. BBa_K5096046 linear DNA Construct of the inhA target construct.

Results

PCR and PCR Cleanup

Before testing the linear DNA constructs provided by Integrated DNA Technologies (IDT), we purified our four inhA sgRNAs and the inhA target sequence using polymerase chain reaction (PCR) kits from Thermo Scientific, then performed PCR cleanup using kits from Qiagen (see Table 1). By using highly purified DNA, we aimed to achieve more accurate results, as variability in purity and yield can influence expression strength and cause inconsistencies between trials, potentially leading to misleading conclusions.

Dilution to create 120 nM and 20 nM stock solutions of sgRNAs and the target construct, respectively, follows the experimental protocol developed that were previously used in GFP CRISPRi testing (see CRISPRi GFP).

Testing inhA Target Construct

After conducting PCR and PCR clean up on our inhA target construct, we tested the diluted inhA target construct (1nM working concentration) by measuring the RFU values of deGFP fluorescence produced in the myTXTL Pro Cell-Free Expression Kit’s Master Mix using various volumes (see Table 2 and Fig. 5).

Figure 5. Graphs showing deGFP expression from the inhA target construct (1nM working concentration) at different volumes. At 2.5uL of the inhA target construct, approximately 200 RFU were produced. At 0.6uL, the expression was around 50 RFU, showing an insignificant difference compared to the negative control.

Troubleshooting Low Fluorescence

We observed around 200 RFU by using 3uL of diluted inhA target construct (1nM). However, the RFU values from using 0.6uL of the diluted target construct were significantly lower than expected, at approximately 50 RFU, showing no significant difference compared to the negative control. Given the myTXTL Pro kit’s maximum volume limit of 12uL, which only accommodates 3uL of reagents apart from the 9uL of master mix, we needed to set aside volumes for additional reagents like dCas9 and sgRNA specific to the CRISPRi reaction. Consequently, to achieve higher RFU values with smaller volumes of the inhA target construct, we increased the working concentration of the inhA target construct by using the initial 100nM stock concentration instead of the diluted 20nM stock solution, which resulted in a five-time increase in working concentration of inhA target construct from 1nM to 5nM. By utilizing 0.6uL of the 100nM stock concentration, we achieved around 150 RFU, which was similar to the values obtained when using 3uL of 20nM diluted stock concentration (see Fig. 6).

Figure 6. Graph of deGFP expression from 0.6uL of 100 nM stock concentration (5nM working) of inhA target construct, resulting in 150 RFU.

Testing inhA sgRNA

Note that concentrations enclosed in parentheses refer to the working concentrations of reagents.

We tested our inhA sgRNAs by adjusting their working concentrations and the inhA target construct to maintain an approximate 1:5 ratio (construct:sgRNA) following the GFP CRISPRi concentration reaction protocol (see CRISPRi GFP). In total, we conducted 11 reaction: a high positive control (5 nM), a regular positive control (1 nM), a negative control, and 8 experimental groups (see Fig. 7).

However, when comparing the unsuccessful sgRNA71 to sgRNA70, we confirmed that sgRNA70 (BBa_K5096071) was effective in repressing the deGFP expression. More specifically, we determined that sgRNA70 was effective when paired with a sgRNA70 (24.75nM) and inhA target construct (5nM) (see Fig. 8). Therefore, we followed the advice from Ms. Kathyrn Eckartt – a PhD student specializing in CRISPRi research at Rockefeller University – to identify sgRNA70’s percent repression. She informed us that at least a 50% decrease in fluorescence was necessary for the functional repression of the inhA gene. After calculating the percent repression of sgRNA70 on the inhA target construct, we found that sgRNA70 achieved a 60.4% decrease in fluorescence compared to the positive control. This significant reduction indicates that sgRNA70 can effectively downregulate the inhA gene, disrupting the pathogenicity of M. tuberculosis and enhancing our ability to eliminate the bacteria.

Figure 7. Successful inhA CRISPRi using sgRNA70 (24.75nM) and inhA target construct (5nM) achieving 60.4% decrease in fluorescence compared to the positive control.

Future

When testing inhA sgRNA69, sgRNA50, and sgRNA71, we did not achieve successful CRISPRi reactions (see CRISPRi Lab Notebook). In contrast, despite the pairing between sgRNA70 (24.75nM) and inhA target construct (5nM) achieving success, the lower concentrations of sgRNA70 (5nM) and the inhA construct (1nM) only yielded RFU values slightly higher than the negative control (see Fig. 8). The unexpected outcome of sgRNAs producing higher fluorescence than the positive control suggests an anomaly in the reaction dynamics, emphasizing the need for continued research and testing as the SHIELD project progresses in the future.

Modeling

Our modeling committee also utilized MATLAB, a platform that enables wetlab committees to simulate various parameters such as target genes and binding coefficients, predicting experimental success. This approach allows the wetlab committees to focus on the most optimal concentrations and configurations, streamlining our experiments and enhancing efficiency. The reciprocal relationship between the modeling and wetlab results facilitates refinement of both the mathematical predictions and experimental design, ultimately improving the accuracy of our M. tb CRISPRi application (see Model CRISPRi).

References